[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <20260108-ffs-dmabuf-ioctl-fix-v1-0-e51633891a81@samcday.com>
Date: Thu, 08 Jan 2026 08:30:19 +1000
From: Sam Day via B4 Relay <devnull+me.samcday.com@...nel.org>
To: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Cc: linux-usb@...r.kernel.org, linux-kernel@...r.kernel.org,
Sam Day <me@...cday.com>, David Heidelberg <david@...t.cz>
Subject: [PATCH 0/2] usb: gadget: f_fs: DMA-BUF fixes
I recently started kicking the tires on DMA-BUF support in FunctionFS.
In the process, I found a couple of issues that are fixed in this
series.
One issue is a minor bug in how the ioctl handler treats invalid calls.
The other is more serious and breaks DMA transfers on systems with a
tenacious SMMU.
I created a fairly minimal reproducer of the DMA issue in this gist, to
assist in testing:
https://gist.github.com/samcday/d23aaf95be678bcdd1dd62b4b79861ea
Tested on:
- Pixel 3a
- OnePlus 6T
Signed-off-by: Sam Day <me@...cday.com>
---
Sam Day (2):
usb: gadget: f_fs: Fix ioctl error handling
usb: gadget: f_fs: fix DMA-BUF OUT queues
drivers/usb/gadget/function/f_fs.c | 24 +++++++++---------------
1 file changed, 9 insertions(+), 15 deletions(-)
---
base-commit: f0b9d8eb98dfee8d00419aa07543bdc2c1a44fb1
change-id: 20251114-ffs-dmabuf-ioctl-fix-272497182f86
Best regards,
--
Sam Day <me@...cday.com>
Powered by blists - more mailing lists